INDEX

Explanations

repetitions of the word "while" and specific contextual keywords related to programming logic

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 anom

-0.49

 dones

-0.49

GetAxis

-0.48

tena

-0.47

Fah

-0.47

auw

-0.46

dsn

-0.46

jez

-0.46

 Naked

-0.45

/><

-0.45

POSITIVE LOGITS

GEBURTSDATUM

0.95

astify

0.84

postIndex

0.84

 ProtoMessage

0.80

endregion

0.80

 setCode

0.78

 siguran

0.76

KURZBESCHREIBUNG

0.73

قایناق‌لار

0.72

ftagPool

0.70

Activations Density 0.032%

repetitions of the word "while" and specific contextual keywords related to programming logic

No Comments

No Known Activations